Orthostatic hypotension (OH) is commonly observed among institutionalized elderly people and is a potential source of morbidity and mortality from falls, syncope, dizziness, fractures, or vital organ ischemia. Despite the potential importance of orthostatic hypotension as a contributor to disability in advanced age, very little information is available about its clinical determinants and relationship to morbid outcomes. The aims of this proposal are to identify 1)clinical determinants and 2) long-term consequences of orthostatic hypotension in an institutionalized elderly population. We will study 6,000 residents of 55 National Health Corporation (NHC) nursing homes located across the south and SE U.S. These homes represent the gender and racial distributions of nursing home residents across the country. Subjects will be long-stay residents who, at the baseline assessment, are willing and able to stand upright for postural BP measurements. Sixty-five skilled nurses serving these homes will undergo one-day group training sessions in orthostatic BP measurement at regional sites where the investigators will provide didactic lectures, hand-out materials, videotapes, practice sessions, interrater reliability tests, and calibrated sphygmomanometers to assure optimal BP measurement technique. These nurses will then obtain 5 within-day and 12 monthly measurements of orthostatic BP and pulse, for each subject. A research nurse will monitor the conduct and reliability of BP measurements twice during the year at 10 randomly chosen sites. In addition to BP data, extensive sociodemographic, clinical, functional, medication, health service utilization, and outcome data will be gathered at each nursing home at quarterly intervals, using well-established instruments and procedures that have been operational at NHC homes for many years. These data will include variables from the HCFA-mandated minimum data set and resident assessment system. All data will be computerized and sent to the HRCA Research and Training Institute at regular intervals, according to previously well-established procedures. The appropriate definition of OH will be determined first, by examining repeated postural BP measurements within a single day, and monthly over a one-year period, to identify discrete patterns of BP response; then, by validating each pattern of response against specific clinical correlates and outcomes over a 4-year followup period. The clinical determinants of various patterns of OH will be identified using multivariate regression techniques. The long-term consequences of OH will be determined using life-table and proportional hazards methodology. The unique availability of a large, well-characterized, and closely followed population of elderly nursing home residents, and the extensive experience of the investigators in long-term care research, blood pressure variability, and clinical epidemiology, will assure that this project achieves its stated goals.